Practical application of fuzzy logic and neural networks to fractured reservoir characterization

Until recently most fractured-reservoir modeling tools were limited to simple discrete statistical models. A new approach in fractured-reservoir characterization which uses artificial intelligence tools is described in this paper. The methodology is based on the assumption that there is a complex relationship between a large number of potential geologic drivers and fractures. Structure, bed thickness and lithology are a few of the drivers that played a role when fractures where created. The first step in the described methodology is the ranking of all existing geologic drivers. A fuzzy neural network is used to evaluate the hierarchical effect of each geologic driver on the fractures. As a result, the geologist or reservoir engineer will be able to identify locally and globally the key geologic drivers affecting fractures. The second step of the approach is to create a set of stochastic models using a backpropagation neural network that will try to quantify the underlying complex relationship that may exist between key geologic drivers and fracture intensity. The training and testing of the neural network is accomplished using existing data. The third step of the approach is to perform an uncertainty analysis by examining the fracture cumulative distribution function resulting from the large number of stochastic models. Using these three steps, the 2D or 3D distribution fractures and their underlying uncertainty can be determined at undrilled locations. The methodology is illustrated with an actual tight gas fractured sandstone. (C) 2000 Elsevier Science Ltd. All rights reserved.